the big thread about cube decisions - Page 8 - Backgammon Forum - Discuss Backgammon Strategy, Clubs and Books

Cube analysis
Rollout cubeless equity  +0,519

Cubeful equities:
1. Double, take         +0,822
2. Double, pass         +1,000  ( +0,178)
3. No double            +0,631  ( -0,191)
Proper cube action: Double, take

Giving black less opportunity to pick up a second checker by moving from 22 to 8:

White - Pips 114

Black - Pips 103

Code:

Cube analysis
Rollout cubeless equity  +0,515

Cubeful equities:
1. Double, take         +0,802
2. Double, pass         +1,000  ( +0,198)
3. No double            +0,637  ( -0,166)
Proper cube action: Double, take

Now there is only one checker left behind, there is more ammunition and a pip lead, which seem to compensate for less gammon chances.

white has entered on 5-point:

White - Pips 109

Black - Pips 117

Code:

Cube analysis
Rollout cubeless equity  +0,140

Cubeful equities:
1. No double            +0,165
2. Double, pass         +1,000  ( +0,835)
3. Double, take         -0,011  ( -0,176)
Proper cube action: No double, take (17,4%)

I think that decisive here is that 5's are useful at both sides of the board.

Black checker from 23 to 24:

White - Pips 114

Black - Pips 118

Code:

Cube analysis
Rollout cubeless equity  +0,321

Cubeful equities:
1. No double            +0,436
2. Double, pass         +1,000  ( +0,564)
3. Double, take         +0,373  ( -0,063)
Proper cube action: No double, take (10,1%)

Less indirect shots and ways to get out, though doubling is no disaster.

Shifting to the 9-point:

White - Pips 114

Black - Pips 114

Code:

Cube analysis
Rollout cubeless equity  +0,650

Cubeful equities:
1. Double, pass         +1,000
2. Double, take         +1,105  ( +0,105)
3. No double            +0,746  ( -0,254)
Proper cube action: Double, pass

If white enters with a 3, there will be two pointers instead of one. This is sufficient to make it a firm pass.

Quote

05-04-2015 , 05:14 AM

#177

yogiman

adept

Join Date: Jun 2012 Posts: 784

Position ID: ZrsAwCPM3Q4gQA Match ID: cAkAAAAAAAAE

White - Pips 155

Black - Pips 119
Black on roll. Cube action?
Created with www.BGdiagram.com

Quote

05-04-2015 , 09:17 AM

#178

Kamba

enthusiast

Join Date: Feb 2015 Posts: 93

Yoiks. Right or wrong, I don't like doubling in these situations for fear that white will get a checker out before I do. If I get my checker off the bar on this throw I'm doubling, even if white gets one guy out of my home.

It looks dangerous for white, but i think if i don't get off the bar now I could easily end up with two checkers on the bar, which would be very bad for me.

So, due to timidity and some not insignificant confusion I would say:

No double / surely a Take.

Quote

05-04-2015 , 10:38 AM

#179

bleep69

grinder

Join Date: Sep 2008 Posts: 483

I think it's a double, despite also having confusion. We need to get out, or in a couple of throws white could either make another point in his home, or hit the pip on 17. However, if we do get out, and we are favoured to do so, we have lots of ammo bearing down on our home, and can build a base rapidly. Coupled with the fact that we are well ahead in the race, I think we are good.

Double/take.

Quote

05-04-2015 , 11:19 AM

#180

Karol Szczerek

enthusiast

Join Date: Mar 2015 Posts: 92

It looks like a no-double. We got many jokers, but also many anti-jokers. Homever, with white having 5(!) checkers behind our prime, which is a lot, it could be a marginal double. I wouldn't double here, although I think the decision might be close.

If You take one more checker from opp's 21pt. and put in in the outfield, I'm pretty sure doubling would be very premature, probably a whooper / double-whooper.

Quote

05-04-2015 , 03:19 PM

#181

peachpie

centurion

Join Date: Aug 2014 Posts: 182

Definitely no double for me. On the bar against a 4 point board, no anchor, no outfield control, op has a good anchor ... the list is really long.

OK, if black escapes right now, he might lose his market, but that is only on 8 rolls. On the rest, white is doing just fine, and has good chances for a recube. In fact, this would be far from the worst beaver I have seen.

Quote

05-05-2015 , 04:50 AM

#182

Taper_Mike

adept

Join Date: Sep 2009 Posts: 1,030

The take must be easy. Black will dance 45% of the time, and afterwards, White will roll the 5 or 6 that allows him to jump into the outfield 55% of the time. In case White does not roll a 5 or 6, he can handle any of the smaller non-doublets relatively easy. White won’t like it, however, if he rolls 44. To a lesser extent, 33 and 22 are also problematic.

I am much less confident about the double. When Black rolls 63, 53, 43, or 61, he will be very happy, but his other entering numbers do not do much for him other than simply enter. Of course, his 16 dancing numbers are even worse. The one thing he does have going for him is a huge racing advantage, but I don’t see that as enough to cube from the bar.

No double/take.

Mike

Quote

05-06-2015 , 08:15 AM

#183

yogiman

adept

Join Date: Jun 2012 Posts: 784

White - Pips 155

Black - Pips 119

Code:

Cube analysis
Rollout cubeless equity  +0,504

Cubeful equities:
1. Double, take         +0,759
2. Double, pass         +1,000  ( +0,241)
3. No double            +0,684  ( -0,075)
Proper cube action: Double, take

The pipdifference is of no importance at all. Decreasing the pipdifference and giving some more flexibility by moving 4w to 19 makes it a no double:

White - Pips 140

Black - Pips 119

Code:

Cube analysis
Rollout cubeless equity  +0,320

Cubeful equities:
1. No double            +0,436
2. Double, pass         +1,000  ( +0,564)
3. Double, take         +0,356  ( -0,080)
Proper cube action: No double, take (12,4%)

I approach this problem by outlining methodically the events that can happen after white's turn:
1. black enters
1.1 white gets into the outfield
1.2 white doesn't get into the outfield

2. black does not enter
2.1 white gets into the outfield
2.2 white doesn't get into the outfield

1. black enters
In case black enters on the 22-point, next turn he will almost certainly double white out. This is in 30% of the cases.
After 35 for black, and rather good 52 for white:

Code:

Cube analysis
Rollout cubeless equity  +0,631

Cubeful equities:
1. Double, pass         +1,000
2. Double, take         +1,052  ( +0,052)
3. No double            +0,839  ( -0,161)
Proper cube action: Double, pass

So what is left is when black enters on the 24-point.

1.1 white gets into the outfield
There is not very much changed. After 15 for black and 52 for white:

Code:

Cube analysis
Rollout cubeless equity  +0,555

Cubeful equities:
1. Double, take         +0,868
2. Double, pass         +1,000  ( +0,132)
3. No double            +0,748  ( -0,120)
Proper cube action: Double, take

This shows that black's situation improves greatly in case he enters. 16 will lead to a pass for white. The probability for this is 3%.

1.2 white doesn't get into the outfield
I presume that white will have a pass. The chance that black throws a 1 is about 30%, and that white doesn't get into the outfield is 45%. That 's in about 15% of the cases.

2. black does not enter
2.1 white gets into the outfield
After 52 for white:

Code:

Cube analysis
Rollout cubeless equity  +0,166

Cubeful equities:
1. No double            +0,189
2. Double, pass         +1,000  ( +0,811)
3. Double, take         -0,035  ( -0,224)
Proper cube action: No double, take (21,7%)

There is a considerable equity loss, but the positive thing is that black is still favorite to win. The chance that black will not enter is 45%, and that white skips the 4-prime is 55%. That's in about 25% of the cases.

2.2 white doesn't get into the outfield
11,21,31,32 will lead to a take, and 22,33,44,41,42,43 will lead to a pass. The probability that white throws one of the bad numbers is 25%, which makes 45% x 25% =11%.

Conclusion:
The passes are all market losers. Question is whether the market losses in total are outweighing the no-doubles in total. I have put the percentages of the passes in red and the no-double in green, showing about 60% passes and 25% no-doubles. Taking in account some too positive judgements, there are twice as many passes. Out of simplicity ignoring the strength of the passes and no-doubles, I think this proofs that this position is a double.

Quote

05-06-2015 , 08:31 AM

#184

Robertie

old hand

Join Date: May 2004 Posts: 1,954

"The pip difference is of no importance at all. Decreasing the pip difference and giving some more flexibility by moving 4w to 19 makes it a no double:"

Aren't you contradicting yourself here?

Quote

05-06-2015 , 10:01 AM

#185

yogiman

adept

Join Date: Jun 2012 Posts: 784

Yes, thank you for correcting me. Still, the question is to what extent white's increase in flexibility is responsible for the no-double. More checkers on the anchor gives more opportunity for containment. 5 checkers behind could also create the opportunity to make the 9w or 10w-point. The equity with less pipdistance has dropped significantly, so it is not likely that that fourth checker on the pile will lead to more gammon losses.

Quote

05-09-2015 , 12:58 PM

#186

yogiman

adept

Join Date: Jun 2012 Posts: 784

Position ID: 8/EBAAfktoHAGA Match ID: cAkAAAAAAAAE

White - Pips 110

Black - Pips 160
Black on roll. Cube action?
Created with www.BGdiagram.com

Quote

05-09-2015 , 01:51 PM

#187

Taper_Mike

adept

Join Date: Sep 2009 Posts: 1,030

As I noted in a previous position, I have a tendency to pass a bit more than I should. Even knowing that, I would pass here. It will be difficult for White make his 4pt and 5pt. Bringing three checkers through the outfield without trading hits at some point will also be difficult. Finding a safe landing spot for the checkers he does manage to bring home will be a further difficulty.

What White does have going for himself is a racing advantage. At any time, he could roll a set of doublets, and solve a large part of his problems in one stroke.

Mike

Quote

05-09-2015 , 02:49 PM

#188

Kamba

enthusiast

Join Date: Feb 2015 Posts: 93

I would drop instantly. I think white is just going to get hit again and again trying to get home.
So, double/drop

Quote

05-09-2015 , 08:38 PM

#189

_Z_

adept

Join Date: Jul 2004 Posts: 722

I would actually scoop this up. With White's big race lead, it going to take Black forever to squash all resistance. White can improve on his position by rolling big doubles, hitting a fly shot or attacking Black as he leaves his anchors. Or just jumping a checker and getting missed. It seems like these possibilities are going to be around for a long time.

White position looks ugly, but there's kind of only one dead checker and he outboards Black 3 to 2 for the time being.

Maybe Black should double based on his positional edge, but I actually think it's better to wait until a threat materializes or White's position deteriorates a little more. From a technical standpoint at least.

No double/take

Last edited by _Z_; 05-09-2015 at 08:45 PM.

Quote

05-09-2015 , 11:24 PM

#190

uberkuber

veteran

Join Date: Jul 2009 Posts: 3,043

As White I would reluctantly take, so I guess it could be a double.

Quote

05-10-2015 , 10:12 AM

#191

Robertie

old hand

Join Date: May 2004 Posts: 1,954

This is a tough position, but I'll go with double/take. Let's start by looking at the take/drop decision.

Black's advantage is clear. He can control the outfield by running off the 20-point, after which we can see the horror sequences for White: White runs into the outfield, gets hit, enters with something like 6-1, splits off his anchor, and gets crushed. That's the bad news.

The good news is that White leads by 50 pips in the race and has, right now, the better board. His 6s and 5s jump and his small 4s don't play, so Black won't kill a lot of checkers over the next few rolls. I think White's anchor, pip lead, and lack of dead checkers means he'll be in this game for a long time in many variations. So I take.

Black's enough of a favorite so the double looks OK from a technical standpoint. As a practical matter, it's hugely correct since many players will pass.

This is a position that really illustrates the power of the hand rollouts that we were forced to use before the bot era. Play this position out a hundred times and watch what happens. White will win more games than you think, and every win will look weird and flukey. But the weird, flukey wins should add up to a take. A 50-pip race lead commands respect.

Quote

05-11-2015 , 08:09 AM

#192

yogiman

adept

Join Date: Jun 2012 Posts: 784

White - Pips 110

Black - Pips 160
Black on roll. Cube action?
Created with www.BGdiagram.com

Code:

Cube analysis
Rollout cubeless equity  +0,539

Cubeful equities:
1. Double, take         +0,888
2. Double, pass         +1,000  ( +0,112)
3. No double            +0,866  ( -0,022)
Proper cube action: Double, take

19w to 22:

White - Pips 107

Black - Pips 160

Code:

Cube analysis
Rollout cubeless equity  +0,598

Cubeful equities:
1. Double, pass         +1,000
2. Double, take         +1,064  ( +0,064)
3. No double            +0,935  ( -0,065)
Proper cube action: Double, pass

Killing a second checker turns out to be to much. If white throws 21, 31, 32, 33 or 43 next turn, black will have lost his market, though slightly. So it is proper for black to double. But taking is still no disaster.

22w to 1:

White - Pips 131

Black - Pips 160

Code:

Cube analysis
Rollout cubeless equity  +0,603

Cubeful equities:
1. Double, pass         +1,000
2. Double, take         +1,053  ( +0,053)
3. No double            +0,930  ( -0,070)
Proper cube action: Double, pass

Indeed does the size of the pips matter.

And there is a lack of containment:

White - Pips 110

Black - Pips 178
[/URL]

Code:

Cube analysis
Rollout cubeless equity  +0,631

Cubeful equities:
1. Double, pass         +1,000
2. Double, take         +1,161  ( +0,161)
3. No double            +0,963  ( -0,037)
Proper cube action: Double, pass

As a matter of fact, apart from white's decision in the original position the cube decisions are technically no big issue here. As mentioned, practically it might be a different matter for black.

Quote

05-14-2015 , 08:53 AM

#193

yogiman

adept

Join Date: Jun 2012 Posts: 784

Position ID: mNcGBDTodgeASA Match ID: cAkAAAAAAAAE

White - Pips 160

Black - Pips 144
Black on roll. Cube action?
Created with www.BGdiagram.com

Quote

05-14-2015 , 09:12 AM

#194

uberkuber

veteran

Join Date: Jul 2009 Posts: 3,043

Double for sure. Small take, but not sure at all for this one.

Quote

05-14-2015 , 12:38 PM

#195

_Z_

adept

Join Date: Jul 2004 Posts: 722

I'd guess that White has enough play to take here, but I'm really not sure. One thing working against him is that Black is very diverse, 2s and 5s hit, 1s, 3s and 4s make the four point. But White's not doing too bad on the hit, hit back sequences. And he's doing fine on Black's bad numbers (most 6s, 44). And he'll always have some sort of game in any case.

I like to say backgammon is a game for optimists...

Double/Take.

Quote

05-14-2015 , 05:07 PM

#196

Taper_Mike

adept

Join Date: Sep 2009 Posts: 1,030

So long as Black does not dance, every number plays well except for 6. With 3 checkers primed already, Black would become a heavy favorite by adding another point to his block with a 1, 3, or 4. With 2 and 5, Black could hit on the 18pt. The double seems clear.

I am less confident about the take. While it is true that Black will usually not both hit and cover, White's prospects look dim. If White gets a 4th checker sent back, his chance of losing a gammon will certainly rise. I am frankly uncertain whether to take or pass. Forced to choose, I would take, and then cross my fingers hoping Black dances or rolls a 6.

Mike

Quote

05-15-2015 , 02:45 AM

#197

bleep69

grinder

Join Date: Sep 2008 Posts: 483

I'm also going with double/take. I think the double is clear, and I say take because white has a lot of potential threats, and could easily end up putting black in a dodgy position with a couple of decent rolls.

Quote

05-17-2015 , 04:33 AM

#198

yogiman

adept

Join Date: Jun 2012 Posts: 784

Code:

Cube analysis
Rollout cubeless equity  +0,510

Cubeful equities:
1. Double, take         +0,773
2. Double, pass         +1,000  ( +0,227)
3. No double            +0,669  ( -0,104)
Proper cube action: Double, take

Hitting white's outfield blot is just an arbitrary move:
Black throws 52:

Code:

    1. bar/20 4/2*                  Eq.:  +0,400
       0,660 0,257 0,024 - 0,340 0,079 0,003 CL  +0,519 CF  +0,400
      
    2. bar/18*                      Eq.:  +0,394 ( -0,006)
       0,655 0,294 0,035 - 0,345 0,104 0,007 CL  +0,527 CF  +0,394

Sixes are bad for black, for example, in case of 63 black is going to hit on the 2-point, and is not anymore favorite to win:

Code:

    1. Rollout          bar/22 8/2*       Eq.:  -0,079
       0,534 0,226 0,029 - 0,466 0,198 0,021 CL  +0,103 CF  -0,079

After the black 63, white has thrown just a moderate 62, hitting from the bar black's checkers on the 2-point and 22-point:

As we expect, black is far removed now from a double:

Code:

Cube analysis
Rollout cubeless equity  +0,127

Cubeful equities:
1. No double            +0,123
2. Double, pass         +1,000  ( +0,877)
3. Double, take         -0,123  ( -0,246)
Proper cube action: No double, take (21,9%)

White's fear will be that black completes his 5-prime, which will happen only half of the time. If he does't make his 5-point alltogether, white has almost 50% chance to do that for him. In the next position, black has completed his 5-prime with 21, followed by a white 41:

Code:

Cube analysis
Rollout cubeless equity  +0,429

Cubeful equities:
1. Double, take         +0,581
2. Double, pass         +1,000  ( +0,419)
3. No double            +0,550  ( -0,031)
Proper cube action: Double, take

Though black has made his 5-prime, he hardly has a double.

So the obvious bright spots for white are that black doesn't complete his 5-prime by throwing a 6 or 25,35,44,55, and that white makes the 20w-point. The chance that one of those happens is about 75%.

Quote

05-17-2015 , 05:02 PM

#199

Taper_Mike

adept

Join Date: Sep 2009 Posts: 1,030

Quote:

Originally Posted by yogiman

Code:

Cube analysis
Rollout cubeless equity  +0,429

Cubeful equities:
1. Double, take         +0,581
2. Double, pass         +1,000  ( +0,419)
3. No double            +0,550  ( -0,031)
Proper cube action: Double, take

Though black has made his 5-prime, he hardly has a double.

By this, I presume you mean that Black has a small double.

In English, there is an idiom by which the use of "hardly" can mean "is not." So if I say, "That's hardly a double," what I am really saying is, "That is not a double." Just to confuse things, "hardly" can also mean "barely," which is the usage I read in your statement. There is an example of this in the Merriam/Webster dictionary: "I hardly knew her," means, "I barely knew her."

Mike

Quote

05-18-2015 , 06:31 AM

#200

yogiman

adept

Join Date: Jun 2012 Posts: 784

Thanks for your linguistic support. Seems I have given you enough reason to doubt me at times. It also suggests that for the rest of the post I am doing pretty okay.

Quote

Page 8 of 13

First

3 4 5 6 7 8 9 10 11 12 13

Last

Post Reply Subscribe

...

Page 8 of 13

First

3 4 5 6 7 8 9 10 11 12 13

Last